Argues PE follows a mechanical template — buy with 70-90% debt, load debt onto the target, extract dividends, exit in 5-7 years — and that this template now controls structurally significant slices of ERs, ambulances, ISPs, and nursing homes. Cites the 2021 NBER paper linking PE-owned nursing homes to a 10% increase in resident mortality as evidence the model causes measurable harm in essential services.
Frames the story as directly relevant to developers because Thoma Bravo, Vista, Insight, and Clearlake have taken private a meaningful share of the enterprise software stack — SolarWinds, Sophos, Proofpoint, ForgeRock, Ping Identity, Anaplan. The same capital structure that hollowed out Toys R Us and Steward Health is now standing behind the security and identity tools engineers rely on.
Senior engineers in the thread describe a recurring lifecycle after their employers were sold to Vista, Thoma Bravo, Insight, or Clearlake: hiring freeze by month three, efficiency review by month nine, support relocated offshore by month eighteen, and a price hike with auto-renewal clauses by year two. The shared complaint is not that PE is profitable but that the profit comes from steady-state degradation of something the buyer never built and doesn't plan to operate past exit.
A Hacker News thread surfacing a long-form piece from Rubbish Talk has reignited a debate developers usually file under 'someone else's problem': private equity now owns a structurally significant share of America's essential services. The article enumerates the surface area — emergency rooms staffed by EmCare and TeamHealth, the two largest ER physician groups, both PE-controlled; ambulance fleets run by Falck and KKR-owned Global Medical Response; fiber ISPs like Ziply (rolled up by Searchlight and WaveDivision out of Frontier's bankruptcy estate); nursing home chains where, per a 2021 NBER paper, PE ownership was linked to a 10% increase in resident mortality.
The pattern is mechanical: a sponsor buys an operating business with 70-90% debt, loads that debt onto the target's balance sheet, extracts dividends, and exits in 5-7 years. The operating company never sees the upside — only the interest expense and the cost cuts required to service it. Hertz, Toys R Us, Red Lobster, Steward Health Care: same template, different industries.
The HN comment thread is more useful than the article itself. Senior engineers in the thread name companies their employers were sold to — Vista, Thoma Bravo, Insight, Clearlake — and describe the same lifecycle: hiring freeze in month three, 'efficiency review' by month nine, the support team relocated to a lower-cost geography by month eighteen, and a price hike with a contract auto-renewal clause by year two. The complaint is not that PE is unprofitable. The complaint is that the profit comes from the steady-state degradation of something the buyer didn't build and doesn't intend to operate past the exit.
For developers this looks like an economics-section story. It isn't. The same capital structure is now standing behind a meaningful slice of the B2B software stack you depend on. Thoma Bravo alone has taken private SolarWinds, Sophos, Proofpoint, ForgeRock, Ping Identity, Anaplan, Coupa, Darktrace, and Imperva over the last six years. Vista owns or has owned Citrix, Tibco, Cvent, Pluralsight, Jamf, KnowBe4. Permira sits behind Zendesk and Genesys. Silver Lake has Qualtrics. Clearlake co-controls Cornerstone OnDemand and Quest Software.
The operational signal arrives on a predictable lag. Quarter one post-close: the public roadmap goes quiet. Quarter two: the senior platform engineers who actually understood the legacy code paths take their RSUs and leave. Quarter three: support response times double, and the status page starts logging incidents that used to be invisible. Quarter four: a 'platform consolidation' announcement migrates customers onto a SKU that costs 30% more for the same feature set. This is the part developers feel directly — the vendor you onboarded for one reason is now operated by people optimizing for a different objective function, and your runbook assumes behavior that no longer holds.
The second-order effect is supply-side concentration. When the same sponsor owns three of the top five identity providers, or both major synthetic-monitoring vendors, the 'multi-vendor' diversification in your DR plan is a paperwork exercise. A 2023 Federal Trade Commission study on PE roll-ups in dermatology, anesthesiology, and dialysis found a consistent 15-30% price increase post-consolidation with no measurable quality improvement. There is no reason to expect the SaaS roll-up math to land somewhere materially different.
The community reaction in the thread is sharper than the article. One frequently-upvoted comment ('we keep treating this as a market problem when it is a regulatory capture problem') captures the consensus: the LBO incentive structure is not a bug in capitalism, it is a feature of the tax code — interest is deductible, dividends are not, so the optimal way to extract value from a stable cash-flow business is to lever it up and pay yourself the spread. Until carried-interest tax treatment and interest-deductibility caps move, the math will keep favoring this structure over operating the business well.
Three concrete moves.
First, when you do vendor due diligence, check the cap table the same way you check the SOC 2. If your critical SaaS is owned by a PE sponsor on year four of a five-year hold, the next 18 months will include a price increase, a forced migration, or a sale to a strategic acquirer who will sunset the product. None of these are hypothetical. Build the exit cost into your TCO model now, not in the renewal cycle.
Second, prefer vendors with durable governance — founder-led, dual-class, employee-owned, or large enough that the LBO math doesn't pencil. Open source with a credible foundation backing it (CNCF, Apache, Linux Foundation) is the cleanest version of this. The reason `etcd`, `Prometheus`, and `Envoy` are safe long-horizon bets is not technical, it is structural: nobody can buy the project and load it with debt.
Third, audit your on-call dependencies for PE concentration. If your incident response routes through PagerDuty (public, but watch the activist pressure), Atlassian (founder-controlled, safe), and a status page vendor owned by Vista, you have a single sponsor of failure in the chain. The fix is not paranoia, it is naming the risk in the same architecture review where you name cloud-region failure.
The PE wave that started in healthcare in the 2010s and moved through fiber and waste hauling in the early 2020s is now mid-cycle in infrastructure software. Expect the same arc: aggressive roll-ups for the next 24-36 months, a wave of price increases as the synergies fail to materialize, and a 2028-2029 reckoning when the debt has to be refinanced into whatever the rate environment looks like then. The developers who win this cycle will be the ones who treat vendor ownership structure as a first-class architectural concern — not the ones who keep being surprised when a perfectly good tool gets quietly worse on a 90-day cadence.
All of the oldest folks I knew were wrong about many things, but they were right to say that morality must be at the center of all public discourse. Pensions were part of the contract of employment for many old companies. As those companies became more interested in ever higher rates of return witho
Huh, that somehow reminds me of Crassus from Rome [1]> The first ever Roman fire brigade was created by Crassus. Fires were almost a daily occurrence in Rome, and Crassus took advantage of the fact that Rome had no fire department, by creating his own brigade—500 men strong—which rushed to burnin
Article doesn't really dig into the angle I personally find most horrifying, strip-mining social capital.In my area PE is gobbling up mom-and-pop apartment complexes, plumbing companies, restaurants, and generally making customers and employees alike pretty miserable.Hard-working founders shoul
I simply don't understand why leveraged buy-out(LBO) is allowed in the first place. It is like paying for the company with the money from the company you are buying.
Top 10 dev stories every morning at 8am UTC. AI-curated. Retro terminal HTML email.
The irony is that PEs exist largely because of pension funds. So to sum it up (not so nicely) we are transferring value from our current standard of living to pay for retirement checks for our old folks.Pensions fund a significant part of PE and they do so because they need around a 7% return in ord